yes, I've always thought formal touch-typing training is better suited to people typing English text than to programming.

I was never formally trained at all. I just typed, mostly on my own time, since an early age, and my motor cortex gradually optimised my actions without me really paying attention, until I could hit >100wpm on a good day. But it's noticeable that the style I evolved is not the usual touch-typing style with a fixed posture on the home keys. It's far more flexible – I can adapt to having a few fingers out of action (e.g. holding a pen or a snack) almost without having to think about it, just using whatever fingers I have for whatever they need to do. And I never use Caps Lock, because I find it easier to hold down Shift with one finger and just use the other 9 however seems sensible. And often I find my left hand covers _more_ than half of the letter keys, while my right hand is free to go for punctuation symbols and the arrow keys (or all the way over to the mouse or trackball), since – as you say – programming is much more heavily weighted towards both weird characters and editing your existing code.

Another noticeable thing about my posture is that my wrists don't bend outwards in the way they'd have to to settle my fingers on the home keys. My hands land on the keyboard at an angle, and just deal with it. Perhaps this is good for getting less RSI!

@Slash909uk oh, wow! I hadn't even begun to consider the influence of non-typing keyboard activities like gaming.

I don't _think_ I have any influences like that – I don't normally play twitchy games on a PC keyboard. (Console controllers more usually.)

But that does remind me that _something_ has happened to my keyboard-related motor control since I was a child. I used to play games on a ZX Spectrum, where if you didn't have a joystick the usual keyboard layout was QAOP for directions and M for fire. Nowadays, if I try to use anything like that layout, I just can't – having the fire control on the same hand as two of the direction keys doesn't work for me, because I can't hammer my thumb for continuous fire and also use two fingers of the same hand for 'steer carefully through these rocks' without neural crossover causing mistakes in both of those actions.

Instead I have to allocate a whole hand to the firing control now, and use a reverse-T of arrow keys on the other hand for movement (whether WASD or the actual arrows). That's objectively less good, it seems to me, because three fingers between four arrow keys isn't enough, and sometimes I die because I can't move my finger between ↑ and ↓ quickly enough. But my motor cortex won't cope with anything else any more!

@dec23k I wrote one of those at university, as part of a prank in which we made a chatbot basically implementing the good old ELIZA algorithm, and sent it on to a bulletin board to wreak havoc. We made it delay a bit before replying, and insert random typos, in the hope that at least _some_ people would be fooled for long enough to be amusing.

Even so, we weren't very subtle about it – we called the account 'eliza'. But that meant it had a female name, in an online chat room, in the mid 1990s, and someone was fooled enough to spend 15 minutes trying to chat it up, in spite of everyone else having clearly figured it out, and _trying_ to tell him!

@fritzoids in speech, one of the really interesting ones to me is the spoonerism. It's easy to see how the _first_ half happens: as you say, you're speaking word #1 while already thinking about word #2 and there's a crossover, so you turn (for example) "test words" into "west ...". But this idea doesn't explain the second half, where the part that went missing from word #1 is hauled back out of some kind of buffer and belatedly emitted. It would seem more natural to me if you _stop_ after saying the mangled first word, suddenly realising you've gone wrong somewhere.

Last summer, at home with my windows open, I overheard an angry mother in the street outside yelling at her child "Go brush your hair and wash your ..." and then she stopped in confusion, because she'd meant to say "wash your hair and brush your teeth", and realised "wash your teeth" didn't make any sense. That's what I _expect_. Continuing on to say the nonsensical second half without even realising, that's more interesting!

@clanger9 for some reason that one happens to me most often in my own name! I'm constantly signing off emails as SImon.

But I suppose that since my best known project also changes the case of the second letter of its name, that's only appropriate!

@maghoff my very favourite one of those is this sequence of shell operations, which I _still_ can't stop doing even though it's such a familiar mistake:

git stash
[do some other thing in this git work tree]
^Y

When I want to save a half-written command line in order to run another command first (e.g. I've suddenly realised I need to cd into the right directory before running the complicated thing I was half way through writing), I'll press ^A^K in bash to cut the whole line into bash's paste buffer; then do my other command; then press ^Y to paste it back on to the next command line and carry on typing.

So ^A^K [...] ^Y is a finger habit for "put this half-done thing aside, and resume it later after doing something else first".

So is "git stash" [...] "git stash pop", in a completely different context.

And I often slip between the two, pressing ^Y when I should have typed "git stash pop", because I made a cognitive error at a higher level than the usual kind of typo, and forget _which_ kind of "resume the half-done thing" I was meant to be doing!

Curiously, never the other way around. ^A^K [...] git stash pop is unheard of. Even when I'm half asleep I know _that's_ too silly.

@ploum

Sorry for the nitpicking: those are homophones. Homonyms have identical spellings.

@simontatham

I’ve only been seeing what’s going on over on the side of the world through posts here on Mastodon…

Are y’all okay? You’ve not yet reached “V for Vendetta” levels of government yet?